A task processing method, apparatus, electronic device, storage medium, and program product

Abstract

This disclosure provides a task processing method, apparatus, electronic device, storage medium, and program product. The method includes: generating a first description text for a first task based on at least one task information in a session using a first intelligent agent; generating multiple second tasks for the first task based on the first description text using the first intelligent agent in response to a task triggering operation for the first task; executing the second tasks using a corresponding first tool invoked by the first intelligent agent to obtain execution information of the second tasks and displaying the execution information in the session; and generating a task processing result for the first task based on the execution information. This effectively improves the accuracy and efficiency of task processing, allows users to understand task progress in real time, and reduces user waiting time.

Description

Technical Field

[0001] This disclosure relates to the field of task processing, and more particularly to a task processing method, apparatus, electronic device, storage medium, and program product. Background Technology

[0002] In task processing technologies, a primary agent typically interprets user input to understand the task intent, breaks down the task, and distributes it to different sub-agents for execution. The primary agent then merges the results from these sub-agents and returns them to the task initiator. Throughout this process, the user is unaware of the task's steps and lacks transparency regarding the processing. Furthermore, the final result takes a considerable amount of time to arrive, resulting in long response times and reduced efficiency. Summary of the Invention

[0003] This disclosure proposes a task processing method, apparatus, electronic device, storage medium, and program product, which at least partially solves the technical problems in related technologies such as the inability to know the task processing process, long response time, and low task processing efficiency.

[0004] In a first aspect, this disclosure provides a task processing method, comprising:

[0005] The first agent generates a first description text for the first task based on at least one task information from the conversation. In response to a task triggering operation for the first task, the first agent generates a plurality of second tasks for the first task based on the first description text. The first intelligent agent invokes the corresponding first tool to execute the second task, obtains the execution information of the second task, and displays the execution information in the session; The task processing result of the first task is generated based on the execution information.

[0006] A second aspect of this disclosure provides a task processing apparatus, comprising: A task module is configured to generate a first description text for a first task based on at least one task information in a session using a first agent; and to generate a plurality of second tasks for the first task based on the first description text using the first agent in response to a task triggering operation for the first task. An execution module is used to utilize the first intelligent agent to call the corresponding first tool to execute the second task and obtain the execution information of the second task; A display module is used to display the execution information in the session; The processing module is used to generate the task processing result of the first task based on the execution information.

[0007] A third aspect of this disclosure provides an electronic device including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor, when executing the computer program, implements the method as described in the first aspect.

[0008] A fourth aspect of this disclosure provides a non-transitory computer-readable storage medium storing computer instructions for causing a computer to perform the method described in the first aspect.

[0009] A fifth aspect of this disclosure provides a computer program product including computer program instructions that, when executed on a computer, cause the computer to perform the method as described in the first aspect.

[0010] As can be seen from the above, the task processing method, apparatus, electronic device, storage medium, and program product provided in this disclosure utilize a first intelligent agent to parse at least one task information from a session to obtain a first description text of a first task, thereby generating multiple second tasks and achieving reasonable refinement and decomposition of tasks. Then, the corresponding first tool is invoked to execute the second tasks, and the execution status of the second tasks is displayed in the session in a timely manner, enhancing the transparency and interactivity of task processing. Finally, the task processing result of the first task is generated based on the execution result. This effectively improves the accuracy and efficiency of task processing, allows users to understand task progress in real time, and reduces user waiting time. Attached Figure Description

[0011] To more clearly illustrate the technical solutions in this disclosure or related technologies, the accompanying drawings used in the description of the embodiments or related technologies will be briefly introduced below. Obviously, the accompanying drawings described below are only embodiments of this disclosure. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0012] Figure 1 This is a schematic diagram of the task processing architecture according to an embodiment of the present disclosure.

[0013] Figure 2 This is a schematic diagram of the hardware structure of an exemplary electronic device according to an embodiment of the present disclosure.

[0014] Figure 3 This is a flowchart illustrating the task processing method according to an embodiment of the present disclosure.

[0015] Figure 4 This is a schematic diagram of a session according to an embodiment of this disclosure.

[0016] Figure 5 This is a schematic diagram of the intelligent agent invocation according to an embodiment of the present disclosure.

[0017] Figure 6 This is a schematic diagram illustrating the execution principle of the second task in an embodiment of this disclosure.

[0018] Figure 7 This is a schematic diagram illustrating the script resource configuration of an embodiment of this disclosure.

[0019] Figure 8 This is a schematic diagram illustrating the execution principle of the second tool according to an embodiment of this disclosure.

[0020] Figure 9 This is a schematic diagram of a task processing apparatus according to an embodiment of the present disclosure. Detailed Implementation

[0021] To make the objectives, technical solutions, and advantages of this disclosure clearer, the following detailed description is provided in conjunction with specific embodiments and the accompanying drawings.

[0022] It should be noted that, unless otherwise defined, the technical or scientific terms used in the embodiments of this disclosure should have the ordinary meaning understood by one of ordinary skill in the art to which this disclosure pertains. The terms "first," "second," and similar terms used in the embodiments of this disclosure do not indicate any order, quantity, or importance, but are merely used to distinguish different components. Terms such as "comprising" or "including" mean that the element or object preceding the word encompasses the elements or objects listed following the word and their equivalents, without excluding other elements or objects. Terms such as "connected" or "linked" are not limited to physical or mechanical connections, but can include electrical connections, whether direct or indirect. Terms such as "upper," "lower," "left," and "right" are used only to indicate relative positional relationships; when the absolute position of the described object changes, the relative positional relationship may also change accordingly.

[0023] It is understood that before using the technical solutions disclosed in the embodiments of this disclosure, users should be informed of the types, scope of use, and usage scenarios of the personal information involved in this disclosure through appropriate means in accordance with relevant laws and regulations, and user authorization should be obtained. For example, in response to receiving a user's active request, a prompt message may be sent to the user to clearly inform the user that the requested operation will require the acquisition and use of the user's personal information. This allows the user to independently choose whether to provide personal information to the software or hardware such as electronic devices, applications, servers, or storage media that perform the operations of the technical solutions of this disclosure, based on the prompt message.

[0024] It is understood that the above notification and user authorization process are merely illustrative and do not constitute a limitation on the implementation of this disclosure. Other methods that comply with relevant laws and regulations may also be applied to the implementation of this disclosure.

[0025] Figure 1 A schematic diagram of an example environment 100 in which embodiments of the present disclosure can be implemented is shown. In this example environment 100, an application 125 is installed on a terminal device 110. A user 140 can interact with the application 125 via the terminal device 110 and / or an attached device of the terminal device 110.

[0026] In some embodiments, application 125 can be downloaded and installed on terminal device 110. In some embodiments, application 125 can also be accessed in other ways, such as through a web page. Figure 1 In environment 100, in response to application 125 being launched, terminal device 110 can display the interface 150 of application 125.

[0027] In some embodiments, terminal device 110 communicates with server 130 to provide services to application 125. Terminal device 110 can be any type of mobile terminal, fixed terminal, or portable terminal, including mobile phones, desktop computers, laptop computers, notebook computers, netbook computers, tablet computers, media computers, multimedia tablets, personal communication system (PCS) devices, personal navigation devices, personal digital assistants (PDAs), audio / video players, digital cameras / camcorders, television receivers, radio receivers, e-book devices, gaming devices, or any combination thereof, including accessories and peripherals of these devices or any combination thereof. In some embodiments, terminal device 110 can also support any type of user-facing interface (such as "wearable" circuitry). Application 125 can be various types of computing systems / servers capable of providing computing power, including but not limited to mainframes, edge computing nodes, computing devices in cloud environments, etc.

[0028] In embodiments of this disclosure, application 125 can provide interaction functionality with an intelligent agent. Application 125 can be an application dedicated to providing intelligent agent services, or an application integrated with an intelligent agent (i.e., it can provide functions or services other than those of an intelligent agent). Although Figure 1 The image shows a single application, but in reality, multiple applications can be installed on the terminal device 110.

[0029] In this disclosure, agents 160-1, 160-2, ..., 160-N (collectively or individually referred to as agent 160) can be deployed locally on terminal device 110 or remotely. In the case of remote deployment, terminal device 110 can directly invoke the agent or invoke the agent via server 130.

[0030] In embodiments of this disclosure, agent 160 may possess intelligent dialogue and task processing capabilities. Terminal device 110 provides an interface 150 that can present interactions with agent 160. In interface 150, user 140 can initiate task requests to agent 160 by inputting natural language (text input or voice input). Alternatively or additionally, user 140 can upload online or offline file conversations to instruct agent 160 to assist in completing various tasks.

[0031] In embodiments of this disclosure, during interaction with user 140, agent 160 can respond to user 140's requests and handle tasks instructed by the user. In some embodiments, during task processing, agent 160 can invoke one or more tools 165-1, 165-2, ... 165-M (collectively or individually referred to as tools 165) to assist in task execution and the provision of task results, as needed for the task. These tools 165 can be any type of tool, such as text generation tools, file reading tools, information search tools, online or offline databases, task processing tools, chart generation tools, web page creation tools, etc.

[0032] In some embodiments, environment 100 may further include a management node for a plurality of agents 160-1, 160-2, ... 160-N, which can interact with the agents 160-1, 160-2, ... 160-N. In some examples, the management node may, in response to a task request from user 140, determine the task requirements corresponding to the task request. The management node may then, based on the task requirements, assign the task request to the agent 160 that matches the task requirements, requesting that agent 160 to execute the task. In other examples, the management node may also determine an execution plan for the task based on the task requirements. The execution plan may indicate one or more subtasks required to complete the task. The management node may assign the one or more subtasks to one or more agents 160, which will then execute their respective subtasks. Regarding the management node, in some examples, the management node may be implemented by one of the agents 160-1, 160-2, ... 160-N (in this case, agent 160 is also referred to as scheduling agent 160). In other examples, management nodes can be implemented using machine learning models, such as language models (LM).

[0033] In some embodiments, agent 160 may be constructed based on one or more machine learning models. In some embodiments, the machine learning model upon which agent 160 is based may include at least a language model (LM), such as a large language model (LLM). In some embodiments, the machine learning model upon which agent 160 is based may include a multimodal model capable of handling multiple modalities of input, such as text input, visual input (e.g., images, videos), audio input, etc. These machine learning models may include content-generating models capable of generating corresponding outputs based on model inputs. In some embodiments, the machine learning model may receive text-modal model inputs (e.g., natural language and / or machine language) and / or non-text-modal model inputs (e.g., images, speech, videos, etc.), and may obtain corresponding model outputs based on model inputs and prompts, thereby completing the task execution.

[0034] It should be understood that the structure and function of the various elements in environment 100 are described for illustrative purposes only and do not imply any limitation on the scope of this disclosure.

[0035] Figure 2 A schematic diagram of the hardware structure of an exemplary terminal device 110 provided in an embodiment of this disclosure is shown. For example... Figure 2 As shown, the terminal device 110 may include: a processor 202, a memory 204, a network module 206, a peripheral interface 208, and a bus 210. The processor 202, memory 204, network module 206, and peripheral interface 208 are interconnected within the terminal device 200 via the bus 210.

[0036] Processor 202 may be a Central Processing Unit (CPU), a Neural Processing Unit (NPU), a Microcontroller (MCU), a programmable logic device, a Digital Signal Processor (DSP), an Application Specific Integrated Circuit (ASIC), or one or more integrated circuits. Processor 202 can be used to perform functions related to the techniques described in this disclosure. In some embodiments, processor 202 may also include multiple processors integrated as a single logic component. For example, such as... Figure 2 As shown, processor 202 may include multiple processors 202a, 202b and 202c.

[0037] Memory 204 can be configured to store data (e.g., instructions, computer code, etc.). Figure 2As shown, the data stored in memory 204 may include program instructions (e.g., program instructions for implementing the task processing method of the embodiments of this disclosure) and data to be processed (e.g., the memory may store configuration documents of other modules, etc.). Processor 202 may also access the program instructions and data stored in memory 204 and execute the program instructions to operate on the data to be processed. Memory 204 may include volatile storage devices or non-volatile storage devices. In some embodiments, memory 204 may include random access memory (RAM), read-only memory (ROM), optical disk, magnetic disk, hard disk, solid-state drive (SSD), flash memory, memory stick, etc.

[0038] Network module 206 can be configured to provide communication with other external devices to terminal device 200 via a network. This network can be any wired or wireless network capable of transmitting and receiving data. For example, the network can be a wired network, a local wireless network (e.g., Bluetooth, WiFi, Near Field Communication (NFC), etc.), a cellular network, the Internet, or a combination thereof. It is understood that the type of network is not limited to the specific examples described above. In some embodiments, network module 206 may include any combination of any number of network interface controllers (NICs), radio frequency modules, transceivers, modems, routers, gateways, adapters, cellular network chips, etc.

[0039] The peripheral interface 208 can be configured to connect the terminal device 200 to one or more peripheral devices to enable information input and output. For example, peripheral devices may include input devices such as keyboards, mice, touchpads, touch screens, microphones, and various sensors, as well as output devices such as displays, speakers, vibrators, and indicator lights.

[0040] Bus 210 can be configured to transfer information between various components of terminal device 200 (e.g., processor 202, memory 204, network module 206, and peripheral interface 208), such as internal buses (e.g., processor-memory bus), external buses (USB port, PCI-E bus), etc.

[0041] It should be noted that although the architecture of the terminal device 200 described above only shows the processor 202, memory 204, network module 206, peripheral interface 208, and bus 210, in specific implementations, the architecture of the terminal device 200 may also include other components necessary for normal execution. Furthermore, those skilled in the art will understand that the architecture of the terminal device 200 described above may only include the components necessary for implementing the embodiments of this disclosure, and does not necessarily include all the components shown in the figures.

[0042] See Figure 3 , Figure 3A schematic flowchart of a task processing method according to an embodiment of the present disclosure is shown. The task processing method according to an embodiment of the present disclosure can be deployed on a server or a terminal. Figure 3 In the process, task processing method 300 may further include the following steps.

[0043] In step S310, the first agent generates a first description text for the first task based on at least one task information in the session.

[0044] The task information can be various media content related to the task input by the user. Task information can include multiple modalities or forms and their combinations. For example, task information can include the user's task requirements expressed in natural language, such as "generate an animation about XX," or it can include user-uploaded images, videos, audio, and other materials, or information mentioned in previous conversations but not yet completed. The first task can be a task determined based on the task information in the conversation; it can be a task with a clear goal and scope parsed from the task information. The first description text is the text content generated after the first agent provides a detailed and accurate description of the first task. The first description text can clearly describe the goal, requirements, and constraints of the first task, providing precise guidance for the breakdown and execution of subsequent tasks, ensuring that each step accurately understands the task requirements. By using the first agent to deeply analyze the task information in the conversation, parsing out the first task, and generating detailed first description text, deviations in subsequent execution caused by inaccurate task understanding are avoided, greatly improving the accuracy of task processing. At the same time, the first description text makes the entire task processing flow more organized and efficient, reducing unnecessary communication and adjustment steps, thereby significantly improving the overall efficiency of task processing.

[0045] Specifically, see Figure 4 , Figure 4 A schematic diagram of a session according to an embodiment of this disclosure is shown. Figure 4In the process, at least two task information entries, 401 and 402, can be input in session 400. Task information 401 can be video clip video_0, and task information 402 can be text information based on natural language description, such as "generate a video like this". A first intelligent agent can be invoked to parse task information 401 and 402, identifying the task intent as generating a similar video based on video clip video_0. To ensure the task processing result more accurately meets user expectations, the first intelligent agent determines that generating a similar video requires specifying the display object in the video clip. Therefore, it can output prompt information 403, "Please upload an image; I can display this image in the video," in session 400 to interact with the client and prompt the user to input more task information. Further task information 404 can be input, which can be an image clip, such as Image, indicating the display object.

[0046] In some embodiments, a first agent generates a first description text for a first task based on at least one task information in a session, including: The first agent receives at least one task information input by the user in the session, parses the task information to obtain a task intent, and generates a first descriptive text to describe the first task based on the task intent.

[0047] In this process, the first intelligent agent identifies the task intent in the task information and generates a first description text for the first task, thus clarifying the task content. This improves the accuracy and efficiency of task processing, reduces inaccurate task results caused by misunderstandings of the task, and meets users' needs for fast and accurate task processing in conversational scenarios.

[0048] like Figure 4 As shown, the first agent can parse task information 401, 402, and 404 to determine that the first task is to generate a video similar to video_0 based on the display object in the image material Image. Furthermore, the first agent can generate a first descriptive text 405 about the first task. This first descriptive text 405 can be displayed in session 400. The first descriptive text 405 can include a brief introduction to the video to be generated, such as theme, purpose, audience, key elements, visual type, sound, aspect ratio, duration, outline, etc.

[0049] In step S320, in response to a task triggering operation for the first task, the first agent generates a plurality of second tasks for the first task based on the first description text.

[0050] Specifically, when a task trigger operation is detected for the first task, the first agent responds to the operation by decomposing and planning the first task based on the first descriptive text. This refines the first task into multiple second tasks with clear objectives and operability. This improves the flexibility and efficiency of task processing, allowing multiple second tasks to proceed in parallel or sequentially, thus shortening the overall task completion time. Simultaneously, the clear objectives of the second tasks reduce execution difficulty and the probability of errors, ensuring that the final task result more accurately meets user needs.

[0051] In some embodiments, the session includes a first session associated with the first agent; In response to a task triggering operation for the first task, the first agent generates multiple second tasks for the first task based on the first description text, including: The first description text is displayed in the first session, and a task triggering control is displayed in the associated area of ​​the first description text; In response to a task triggering operation on the task triggering control, the first agent decomposes the first task based on the first description text to generate multiple second tasks.

[0052] In the first session associated with the first intelligent agent, a first descriptive text can be displayed, and a task trigger control can be set in its associated area. This allows users to intuitively understand the core content of the task and clearly define the operation entry point. When the user interacts with the trigger control, the first intelligent agent breaks down the first task into multiple second tasks based on the first descriptive text.

[0053] like Figure 4 As shown, an editing control 406 can also be displayed in the associated area of ​​the first description text 405 in session 400 for editing the first description text 405. Users can adjust the task description in a timely manner according to the progress of the session or other relevant information, ensuring that the description always accurately reflects their intent. This enhances the smoothness of interaction between the user and the system, allowing users to experience the system's rapid response to their needs. It also simplifies the operation process, improves the intuitiveness and convenience of operation, and conforms to the user's natural usage habits. A task processing control 407 can also be displayed in the associated area of ​​the first description text 405 in session 400 to trigger the execution of the first task. The task processing control 407 can display a control name, which can be determined based on the first task. For example, if the first task is to generate a video, it can display "Generate". A trigger operation (such as clicking) can be performed on the task processing control 407, displaying the text 408 "Generate Video" in session 400 to trigger the execution of the first task; alternatively, the text 408 "Generate Video" can be entered in session 400 to trigger the execution of the first task.

[0054] The first intelligent agent can break down the first task into multiple second tasks, such as second task 1 "configure basic video settings, including aspect ratio and background music", second task 2 "generate storyboard", second task 3 "generate voice-over", second task 4 "generate visual material for each shot in the video", and second task 5 "package the complete video".

[0055] In step S330, the first agent calls the corresponding first tool to execute the second task, obtains the execution information of the second task, and displays the execution information in the session.

[0056] The first intelligent agent, based on its intelligent matching and scheduling capabilities, can invoke a first tool adapted to the second task to execute it. Execution information during the second task's execution is displayed in the session in real time, allowing the user to monitor the task's progress. This enables users to stay informed about the task's progress, reduces waiting time, and optimizes the user experience.

[0057] In some embodiments, the first agent invokes a corresponding first tool to execute the second task, and obtains the execution result of the second task, including: The first agent invokes the first tool to create the second session, and the input parameters of the second task are passed to the second agent via the second session; The second agent invokes the corresponding second tool to execute the execution steps of the second task based on the input parameters, generates the execution result, and terminates the second session. The execution result is sent to the first intelligent agent by invoking the first tool using the first intelligent agent.

[0058] In this process, the first agent first calls the first tool to create a second session, and then passes the input parameters of the second task to the second agent through this session, establishing a communication path for task execution. The second agent then calls the second tool to perform the execution steps based on the input parameters, generating results and completing the second task. After completion, the second session can be closed to avoid resource waste, and the execution results of the second task are returned to the first agent. This allows for collaborative work among multiple agents through client sessions, ensuring a standardized and orderly task execution process and accurate feedback of execution results, significantly improving the accuracy, efficiency, and resource utilization of task processing. Due to the client's routing function, it connects the related paths between the second agents during operation, without the first agent needing to know the specific existence or operational details of the second agents. When the second agent generates process information while executing the second task, the client organizes and arranges it according to preset rules and user needs through the session, then quickly presents it to the user in an intuitive and easy-to-understand way. This improves the flexibility and scalability of task processing; the first agent does not need to adjust its architecture due to the addition, removal, or change of the second agent, reducing system coupling. Meanwhile, the client quickly displays the execution process information of the second intelligent agent, enabling users to keep track of task progress in real time. This enhances the real-time nature and transparency of the interaction, providing users with a smoother and more efficient user experience, and helps improve user satisfaction and trust in the system.

[0059] Specifically, see Figure 5 , Figure 5 A schematic diagram of an agent invocation according to an embodiment of the present disclosure is shown. Figure 5 In this process, the first agent invokes a second tool (including the second agent) to create a new second session, and sends the input parameters (input parameters) to the second agent via the second session through a second task intent. The second agent then invokes different functional tools A and B to execute the corresponding execution steps and obtain the execution results; and invokes tool C to end the second session. The execution result of the second task is then returned to the first agent as an output parameter (output parameter).

[0060] In some embodiments, the first tool includes a second agent; the session also includes a second session associated with the second agent. The execution information of the second task is then displayed in the session, including: displaying the execution information in the second session; the execution information includes execution progress and / or execution result.

[0061] A tool can be invoked by multiple agents, depending on whether the tool is registered with an agent; information between tools is isolated. Capabilities that can be invoked by an agent (such as functional modules) can all serve as tools for that agent, including other agents.

[0062] The first tool may include a second intelligent agent. Once the second intelligent agent processes the second task and obtains the corresponding execution information (such as execution progress and / or execution results), it directly displays this information in the corresponding second session, allowing the client to obtain the task execution status within the session context of interacting with the second intelligent agent. It should be understood that the main subject of the dialogue in the first session is the first intelligent agent, while the main subject of the dialogue in the second session is the second intelligent agent. Users can directly understand the execution progress or results through the client in the second session, improving the efficiency and convenience of information acquisition.

[0063] like Figure 4 As shown, during the execution of the second task by the second intelligent agent, corresponding execution information can be displayed in session 400. For example, when executing the second task 1, execution information 409 "I am generating your video. First, I will set the basic settings of the video, including aspect ratio and background music." When executing the second task 2, execution information 410 "Now, I am generating a storyboard based on your profile..." can be displayed. When executing the second task 3, execution information 411 "Now, I am generating voice-over." When executing the second task 4, execution information 412 "I am generating visual materials for each shot in the video," and the corresponding execution results 413 shot image1 and image2, ... can be displayed. When executing the second task 5, execution information 414 "I am intelligently packaging the complete video for you," and the task processing result 415 video_1 of the first task obtained based on the execution information of the second tasks 1-5 can be displayed.

[0064] In some embodiments, the second agent invokes a corresponding second tool to execute the execution steps of the second task based on the input parameters, generating the execution result, including: Based on the mapping relationship between the second tool and the execution action, the execution script corresponding to the second tool is determined; Based on the input parameters and the execution script, the execution actions of the execution steps are generated; The second intelligent agent invokes the second tool to execute the execution action, and the execution result is obtained.

[0065] The process involves generating a corresponding execution script based on the mapping relationship between the second tool and the action. This script, combined with input parameters, generates specific execution actions, transforming the abstract second task into actionable tasks. The second agent then invokes the second tool to execute these actions, yielding the results. This approach requires only one set of execution script parsing on both the server and client sides. When converting the output of the second tool into execution actions, the client doesn't need to pre-configure parsing rules or preset tools. After understanding the output of the second tool, the mapping logic between the second tool and the execution actions can be configured, and then the corresponding execution script can be generated. This reduces redundant development work and avoids the maintenance complexity and inconsistencies caused by multiple parsing rules. It also reduces the workload and error probability associated with rule updates and synchronization, lowering the overall system maintenance cost. Furthermore, it allows for more flexible adaptation to different second tools. When a new second tool is added or an existing tool is upgraded, no large-scale client modifications are required; only the mapping logic needs adjustment, improving the system's scalability and adaptability to new business needs.

[0066] See Figure 6 , Figure 6 A schematic diagram illustrating the execution principle of the second task according to an embodiment of the present disclosure is shown. Figure 6 In this process, input parameters are passed to the second agent. These parameters may include specific task requirements, initial data, or other relevant information. Based on the input parameters or internal logic, the second agent selects and invokes "Tool A," "Tool B," and "Tool C" to perform different functions. When a tool (e.g., "Tool A") needs to perform a specific operation, it pulls or downloads the corresponding script (e.g., "Script A") from a script distribution network (e.g., a script Content Delivery Network, or script CDN) based on the "tool_name" (e.g., "Tool A"). The script CDN can be a repository storing multiple scripts, such as "Script A," "Script B," and "Script C." The downloaded script (e.g., "Script A") is dynamically parsed and executed as "actions." After the script execution is complete, an execution result "lv_draft" is generated, which is then returned to the first agent. The second agent's ability to dynamically invoke appropriate tools and scripts based on different needs improves the flexibility and scalability of task processing.

[0067] In some embodiments, determining the execution script corresponding to the second tool based on the mapping relationship between the second tool and the action includes: The resource identifier of the execution script is determined based on the second tool and the current version identifier; The corresponding execution script is obtained from the script resource library based on the resource identifier.

[0068] Specifically, the resource identifier of the corresponding execution script is determined based on the second tool and its current version identifier. The corresponding execution script is then retrieved from the script resource repository based on this resource identifier. This method accurately distinguishes execution scripts corresponding to different versions of the tool, avoiding script confusion caused by tool version updates and ensuring that the correct and compatible execution script is obtained. It achieves centralized management and efficient invocation of scripts, improving resource utilization and retrieval speed. It also facilitates unified maintenance and updates of scripts, enhancing system stability and maintainability.

[0069] Specifically, see Figure 7 , Figure 7 A schematic diagram of script resource configuration according to an embodiment of the present disclosure is shown. Figure 7 This demonstrates the process of obtaining and setting the execution script for the corresponding tool version based on the client version. It can receive the "tool identifier (tool_name B)" from the second tool. Based on the input tool identifier, the client version is determined to inform the subsequent process flow. Depending on the client version (version 1 or version 2), the process will enter different branches. If the client version is version 1, the process enters the "version 1" node. The tool identifiers corresponding to version 1 are "tool identifier A, tool identifier B, tool identifier C". Based on the selected tool identifier, the corresponding resource identifiers are further determined; the resource identifiers corresponding to version 1 are "resource identifier A, resource identifier B, resource identifier C". If the client version is version 2, the process enters the "version 2" node. Based on the selected tool identifier for version 2, the corresponding resource identifiers "resource identifier A', resource identifier B', resource identifier C'" are set. This determines the resource identifier B corresponding to the second tool and the client version, and the script resource B obtained from the script resource library is downloaded.

[0070] Configuring execution script resources can be done similarly to configuring special effects resources. Resources are configured in the script resource library and then downloaded via the general asset SDK. The execution script is treated as a special effects resource, with its identifier (ID) used to request downloading it locally via the asset SDK and loading it into memory as an effects object. The path to the effects object and the tool identifier are used as the file path to directly retrieve the corresponding execution script.

[0071] In some embodiments, based on the input parameters and the execution script, the execution actions of the execution steps are generated, including: The second agent invokes the corresponding second tool, and performs a first execution step based on the input parameters and the first function of the second tool to obtain the result of the first execution step. Based on the execution script and the result of the first execution step, the corresponding execution action is generated; And, by using the second intelligent agent to invoke the second tool to execute the execution action and obtain the execution result, including: The execution action is performed based on the action function of the second tool to obtain the execution result.

[0072] In this process, the second agent invokes the second tool, which uses its first function in conjunction with input parameters to execute the first execution step, obtaining a preliminary result. Then, based on the execution script and this result, it generates specific execution actions. Finally, the second agent invokes the second tool again, using its action function to execute the generated actions, thus obtaining the final execution result. This process ensures standardization and consistency, improves the accuracy and efficiency of task execution, and enables more stable and reliable completion of various task processing tasks.

[0073] See Figure 8 , Figure 8 A schematic diagram illustrating the execution principle of a second tool according to an embodiment of this disclosure is shown. Figure 8 In the process, the input parameters are passed to the second intelligent agent, which may rely on the platform to perform calculations or call resources. Subsequently, the second intelligent agent calls the second tool (initiated by the client). The server of this tool uses an algorithm to first generate video products through the "AI video generation" function, then performs "video layout" and produces layout information. After that, the information is transmitted to the client, which first assembles the layout information into a draft through "action conversion", and finally completes "action landing". The entire process realizes the complete operation from input to video generation and landing.

[0074] In step S340, the task processing result of the first task is generated based on the execution result.

[0075] In this invention, after receiving the execution results of the second task (including execution by a tool invoked by the second agent), the client can integrate, analyze, and transform the execution results based on the logical relationship between the first and second tasks, as well as preset rules and algorithms. This generates a task processing result that matches the first task, achieving organic connection between task results at different levels. In traditional technologies, the first agent merges the task results of the second task and returns them to the initiator, resulting in the first agent bearing too many functional responsibilities and requiring a large platform to handle the entire instruction execution process, which can easily lead to illusions. This invention allows the client to perform the task merging process, better controlling the abnormal situations caused by illusions in the first agent (such as a language model).

[0076] It should be noted that the method of this embodiment can be executed by a single device, such as a computer or server. The method of this embodiment can also be applied in a distributed scenario, where multiple devices cooperate to complete the task. In such a distributed scenario, one of these devices may execute only one or more steps of the method of this embodiment, and the multiple devices will perform task processing together to complete the method described.

[0077] It should be noted that the above description describes some embodiments of this disclosure. Other embodiments are within the scope of the appended claims. In some cases, the actions or steps recorded in the claims can be performed in a different order than that shown in the above embodiments and still achieve the desired result. Furthermore, the processes depicted in the drawings do not necessarily require a specific or sequential order to achieve the desired result. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0078] Based on the same technical concept, corresponding to any of the above embodiments, this disclosure also provides a task processing apparatus, see [link to relevant documentation]. Figure 9 The task processing device includes: A task module is configured to generate a first description text for a first task based on at least one task information in a session using a first agent; and to generate a plurality of second tasks for the first task based on the first description text using the first agent in response to a task triggering operation for the first task. An execution module is used to utilize the first intelligent agent to call the corresponding first tool to execute the second task and obtain the execution information of the second task; A display module is used to display the execution information in the session; The processing module is used to generate the task processing result of the first task based on the execution information.

[0079] In some embodiments, a first agent generates a first description text for a first task based on at least one task information in a session, including: The first agent receives at least one task information input by the user in the session, parses the task information to obtain a task intent, and generates a first descriptive text to describe the first task based on the task intent.

[0080] In some embodiments, the session includes a first session associated with the first agent; In response to a task triggering operation for the first task, the first agent generates multiple second tasks for the first task based on the first description text, including: The first description text is displayed in the first session, and a task triggering control is displayed in the associated area of ​​the first description text; In response to a task triggering operation on the task triggering control, the first agent decomposes the first task based on the first description text to generate multiple second tasks.

[0081] In some embodiments, the first tool includes a second agent; the session also includes a second session associated with the second agent. The execution information of the second task is then displayed in the session, including: displaying the execution information in the second session; the execution information includes execution progress and / or execution result.

[0082] In some embodiments, the first agent invokes a corresponding first tool to execute the second task, and obtains the execution result of the second task, including: The first agent invokes the first tool to create the second session, and the input parameters of the second task are passed to the second agent via the second session; The second agent invokes the corresponding second tool to execute the execution steps of the second task based on the input parameters, generates the execution result, and terminates the second session. The execution result is sent to the first intelligent agent by invoking the first tool using the first intelligent agent.

[0083] In some embodiments, the second agent invokes a corresponding second tool to execute the execution steps of the second task based on the input parameters, generating the execution result, including: Based on the mapping relationship between the second tool and the execution action, the execution script corresponding to the second tool is determined; Based on the input parameters and the execution script, the execution actions of the execution steps are generated; The second intelligent agent invokes the second tool to execute the execution action, and the execution result is obtained.

[0084] In some embodiments, determining the execution script corresponding to the second tool based on the mapping relationship between the second tool and the action includes: The resource identifier of the execution script is determined based on the second tool and the current version identifier; The corresponding execution script is obtained from the script resource library based on the resource identifier.

[0085] In some embodiments, based on the input parameters and the execution script, the execution actions of the execution steps are generated, including: The second agent invokes the corresponding second tool, and performs a first execution step based on the input parameters and the first function of the second tool to obtain the result of the first execution step. Based on the execution script and the result of the first execution step, the corresponding execution action is generated; And, by using the second intelligent agent to invoke the second tool to execute the execution action and obtain the execution result, including: The execution action is performed based on the action function of the second tool to obtain the execution result.

[0086] For ease of description, the above apparatus is described in terms of its functions, divided into various modules. Of course, in implementing this disclosure, the functions of each module can be implemented in one or more software and / or hardware.

[0087] The apparatus of the above embodiments is used to implement the corresponding task processing method in any of the foregoing embodiments, and has the beneficial effects of the corresponding method embodiments, which will not be repeated here.

[0088] Based on the same technical concept, corresponding to the methods of any of the above embodiments, this disclosure also provides a non-transitory computer-readable storage medium that stores computer instructions for causing the computer to perform the task processing method as described in any of the above embodiments.

[0089] The computer-readable medium of this embodiment includes permanent and non-permanent, removable and non-removable multimedia, and information storage can be implemented by any method or technology. Information can be computer-readable instructions, data structures, program modules, or other data. Examples of computer storage media include, but are not limited to, phase-change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technologies, CD-ROM, digital versatile optical disc (DVD) or other optical storage, magnetic tape, magnetic disk storage or other magnetic storage devices, or any other non-transfer medium that can be used to store information accessible by a computing device.

[0090] The computer instructions stored in the storage medium of the above embodiments are used to cause the computer to execute the task processing method as described in any of the above embodiments, and have the beneficial effects of the corresponding method embodiments, which will not be repeated here.

[0091] Based on the same inventive concept, corresponding to the task processing method of any of the above embodiments, this disclosure also provides a computer program product, which includes computer program instructions. In some embodiments, when the computer program instructions are run on a computer, the computer performs each step of each embodiment of the task processing method. Corresponding to the execution entity corresponding to each step in each embodiment of the task processing method, the processor executing the corresponding step may belong to the corresponding execution entity.

[0092] The computer program products of the above embodiments are used to cause the processor to execute the task processing method as described in any of the above embodiments, and have the beneficial effects of the corresponding method embodiments, which will not be repeated here.

[0093] Those skilled in the art should understand that the discussion of any of the above embodiments is merely exemplary and is not intended to imply that the scope of this disclosure (including the claims) is limited to these examples; within the framework of this disclosure, the technical features of the above embodiments or different embodiments can also be combined, the steps can be implemented in any order, and there are many other variations of different aspects of the embodiments of this disclosure as described above, which are not provided in detail for the sake of brevity.

[0094] Additionally, to simplify the description and discussion, and to avoid obscuring the embodiments of this disclosure, the provided drawings may or may not show well-known power / ground connections to integrated circuit (IC) chips and other components. Furthermore, the apparatus may be shown in block diagram form to avoid obscuring the embodiments of this disclosure, and this also takes into account the fact that the details of implementation of these block diagram apparatuses are highly dependent on the platform on which the embodiments of this disclosure will be implemented (i.e., these details should be fully understood by those skilled in the art). While specific details (e.g., circuits) have been set forth to describe exemplary embodiments of this disclosure, it will be apparent to those skilled in the art that the embodiments of this disclosure can be implemented without these specific details or with variations thereof. Therefore, these descriptions should be considered illustrative rather than restrictive.

[0095] Although this disclosure has been described in conjunction with specific embodiments thereof, many substitutions, modifications, and variations of these embodiments will be apparent to those skilled in the art from the foregoing description. For example, other memory architectures (e.g., dynamic RAM (DRAM)) may be used with the embodiments discussed.

[0096] This disclosure is intended to cover all such substitutions, modifications, and variations that fall within the broad scope of the appended claims. Therefore, any omissions, modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this disclosure should be included within the scope of protection of this disclosure.

Claims

1. A task processing method, comprising: The first agent generates a first description text for the first task based on at least one task information from the conversation. In response to a task triggering operation for the first task, the first agent generates a plurality of second tasks for the first task based on the first description text. The first intelligent agent invokes the corresponding first tool to execute the second task, obtains the execution information of the second task, and displays the execution information in the session; The task processing result of the first task is generated based on the execution information.

2. The method according to claim 1, wherein, Using at least one task information from the session, the first agent generates a first description text for the first task, including: The first agent receives at least one task information input by the user in the session, parses the task information to obtain a task intent, and generates a first descriptive text to describe the first task based on the task intent.

3. The method according to claim 1, wherein, The session includes a first session associated with the first agent; In response to a task triggering operation for the first task, the first agent generates multiple second tasks for the first task based on the first description text, including: The first description text is displayed in the first session, and a task triggering control is displayed in the associated area of ​​the first description text; In response to a task triggering operation on the task triggering control, the first agent decomposes the first task based on the first description text to generate multiple second tasks.

4. The method according to claim 1, wherein, The first tool includes a second agent; the session also includes a second session associated with the second agent; The execution information of the second task is then displayed in the session, including: displaying the execution information in the second session; the execution information includes execution progress and / or execution result.

5. The method according to claim 4, wherein, The first intelligent agent invokes the corresponding first tool to execute the second task, and obtains the execution result of the second task, including: The first agent invokes the first tool to create the second session, and the input parameters of the second task are passed to the second agent via the second session; The second agent invokes the corresponding second tool to execute the execution steps of the second task based on the input parameters, generates the execution result, and terminates the second session. The execution result is sent to the first intelligent agent by invoking the first tool using the first intelligent agent.

6. The method according to claim 5, wherein, The second agent invokes the corresponding second tool to execute the execution steps of the second task based on the input parameters, generating the execution result, including: Based on the mapping relationship between the second tool and the execution action, the execution script corresponding to the second tool is determined; Based on the input parameters and the execution script, the execution actions of the execution steps are generated; The second intelligent agent invokes the second tool to execute the execution action, and the execution result is obtained; Specifically, determining the execution script corresponding to the second tool based on the mapping relationship between the second tool and the action includes: The resource identifier of the execution script is determined based on the second tool and the current version identifier; The corresponding execution script is obtained from the script resource library based on the resource identifier.

7. The method according to claim 5, wherein, Based on the input parameters and the execution script, the execution actions of the execution steps are generated, including: The second agent invokes the corresponding second tool, and performs a first execution step based on the input parameters and the first function of the second tool to obtain the result of the first execution step. Based on the execution script and the result of the first execution step, the corresponding execution action is generated; And, by using the second intelligent agent to invoke the second tool to execute the execution action and obtain the execution result, including: The execution action is performed based on the action function of the second tool to obtain the execution result.

8. A task processing apparatus, comprising: The task module is used to generate a first description text of a first task based on at least one task information in the session using the first intelligent agent. In response to a task triggering operation for the first task, the first agent generates a plurality of second tasks for the first task based on the first description text. An execution module is used to utilize the first intelligent agent to call the corresponding first tool to execute the second task and obtain the execution information of the second task; A display module is used to display the execution information in the session; The processing module is used to generate the task processing result of the first task based on the execution information.

9. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor, when executing the computer program, implements the method as claimed in any one of claims 1 to 7.

10. A non-transitory computer-readable storage medium storing computer instructions for causing a computer to perform the method according to any one of claims 1 to 7.

Images